DOJ Charges Tech CEO in $421M AI Washing Fraud

What happened

On paper, iLearningEngines was a Silicon Valley fairy tale. A Maryland-based AI company founded in 1999, it had reinvented itself during the artificial-intelligence boom as an “AI-driven business automation platform” that helped organizations train employees, manage compliance, and optimize operations. In April 2024, it went public on the Nasdaq under the ticker AILE, and investors briefly valued the company at roughly $1.5 billion.

According to a 10-count federal indictment unsealed on April 17, 2026, the fairy tale was a complete fabrication. Federal prosecutors allege that iLearningEngines was, from roughly January 2019 through April 2025, a “systemic fraud” built on sham contracts, fake customers, and circular cash flows designed to deceive investors, lenders, and auditors.

The founder and CEO, Puthugramam “Harish” Chidambaran, 57, was arrested in Potomac, Maryland. The former CFO, Sayyed Farhan Ali “Farhan” Naqvi, 44, was arrested in San Jose, California. Both face charges that include running a continuing financial-crimes enterprise, a count that carries a mandatory minimum of 10 years in prison and a maximum of life.

The case represents a significant escalation in federal enforcement against what regulators call “AI washing”: the practice of overstating a company's artificial-intelligence capabilities or performance to attract investors, customers, or media attention. It signals that the Department of Justice now considers AI-related securities fraud a top-tier prosecution priority, alongside traditional financial crimes.

What the indictment alleges: a web of sham contracts

The 2026 indictment, filed in federal court in Brooklyn by the U.S. Attorney's Office for the Eastern District of New York, paints a devastating picture of a company that prosecutors say was fraudulent from top to bottom.

According to the charging documents, iLearningEngines reported $421 million in revenue for 2023, a figure that would have made it a legitimate mid-cap AI success story. Prosecutors allege that at least 90% of that revenue was fabricated. The company did not have tens of millions of dollars in real customer contracts. It had an “intricate web” of fake agreements with shell entities that were secretly controlled by iLearning employees, associates, friends, and even family members.

1. Sham contracts with controlled entities

The indictment alleges that Chidambaran and Naqvi created or co-opted dozens of shell companies that posed as legitimate iLearning customers. These entities signed contracts for iLearning's AI products, generating the paperwork that auditors and investors rely on. But the contracts were not real. The “customers” had no intention of using the products, and in many cases, they were not independent businesses at all, they were alter egos of the executives themselves.

2. Round-trip cash transfers

A sham contract is not enough to fool sophisticated auditors. The money must move. Prosecutors allege that iLearning used investor funds and lender proceeds to make “payments” to these shell customers, which then routed the same money back to iLearning as revenue. This circular flow, known as a “round-trip” transaction, creates the illusion of organic sales while simply recycling the company's own cash.

In practice, a round-trip works like this: iLearning wires $10 million to a shell customer. The shell customer, controlled by an iLearning insider, then “pays” iLearning $10 million for AI software. On iLearning's books, that $10 million appears as revenue. In reality, no sale occurred, no product changed hands, and the company is simply moving its own money in a circle.

3. Related-party concealment

To prevent auditors from connecting the dots, the indictment alleges that Chidambaran and Naqvi concealed the relationships between iLearning and its shell customers. They did not disclose in SEC filings that major “customers” were controlled by employees, friends, or family. They allegedly provided false explanations for large cash flows and fabricated supporting documentation to make sham transactions look legitimate.

4. Lending fraud and investor deception

The inflated revenue numbers did more than pump the stock price. According to the DOJ, iLearning used its fraudulent financial statements to secure tens of millions of dollars in loans from lenders who relied on the company's reported performance. Those lenders would not have extended credit had they known that 90% of the company's revenue was fake.

Investors who bought AILE stock on the Nasdaq also relied on the company's public filings. When the truth began to emerge in August 2024, the stock collapsed. By December 2024, iLearning had filed for Chapter 11 bankruptcy in the District of Delaware. In 2025, the case converted to a Chapter 7 liquidation, meaning the company is being dissolved and its assets sold to pay creditors.

The quote that defines the case

In the DOJ's press release announcing the arrests, United States Attorney Joseph Nocella, Jr., delivered a line that neatly encapsulates the government's theory of the case:

“While the defendants pitched iLearning as a way to revolutionize training and education through AI, the truly artificial part of the defendants' story was iLearning's customers and revenues.”

The double meaning is deliberate. The company sold “artificial intelligence,” but the indictment alleges that the only artificial things were the customer relationships and the revenue figures.

The collapse: from $1.5 billion to Chapter 7

• April 2024: iLearningEngines goes public on the Nasdaq via a SPAC merger. Market capitalization peaks near $1.5 billion.
• August 2024: A short-selling research firm publishes a report alleging that iLearning has materially misrepresented its revenue and that many of its announced customers are shell entities. The stock begins to fall.
• September–November 2024: Major investors dump shares. The board launches an internal investigation. Lenders demand repayment.
• December 2024: iLearningEngines files for Chapter 11 bankruptcy in the District of Delaware.
• 2025: The bankruptcy case converts to Chapter 7 liquidation. The company ceases operations.
• April 17, 2026: The DOJ unseals its indictment. Chidambaran and Naqvi are arrested. The government announces it will seek forfeiture of all proceeds from the alleged fraud.

Shareholders who bought AILE at its peak lost essentially everything. The only remaining value is the possibility of recovering pennies on the dollar through securities class-action lawsuits.

Why it matters

The iLearningEngines indictment is not the first AI washing case, but it is by far the largest and most visible. It arrives alongside a growing body of federal enforcement targeting companies that have used the AI hype cycle to defraud investors.

The Nate precedent: when “AI” meant human call centers

In 2025, the DOJ charged Albert Saniger, founder of the e-commerce app nate, with securities fraud. According to prosecutors, Saniger had told investors that nate's checkout technology was powered by “proprietary AI” that could automatically complete purchases across thousands of retail websites. In reality, the company relied on human workers in call centers in the Philippines and Romania to manually complete transactions.

The nate case was a warning shot: the DOJ will scrutinize claims about AI capabilities, and if the technology does not match the marketing, executives can face criminal charges. The iLearningEngines case takes that enforcement to a different scale, not just overstating what the AI could do, but fabricating the underlying business entirely.

The definition of AI washing

The term “AI washing” is a deliberate echo of “greenwashing”, the practice of overstating environmental credentials. In both cases, companies capitalize on a socially desirable label (eco-friendly, AI-powered) to attract capital and customers without delivering the underlying substance.

• Capability washing: claiming that a product uses AI when it relies on simple rules engines or human labor.
• Performance washing: exaggerating the accuracy, speed, or sophistication of a real AI system.
• Revenue washing: what iLearning is accused of, fabricating customer adoption and revenue tied to AI products.
• Pipeline washing: announcing AI products that do not yet exist to boost stock prices.

Why the DOJ is making an example of this case

Legal observers have noted that the iLearningEngines indictment contains several unusual features that suggest the DOJ is treating this case as a template for future AI fraud prosecutions.

1. The continuing financial-crimes enterprise count. Most securities fraud cases are charged under statutes that carry 20- or 25-year maximum sentences. The iLearning indictment includes a count under 18 U.S.C. § 225, which carries a mandatory minimum of 10 years and a maximum of life. This is the same statute used against major money launderers and leaders of large-scale fraud rings. Its use here signals that the DOJ views the alleged scheme as an organized criminal enterprise, not merely white-collar overreach.
2. The 10-count structure. The indictment charges Chidambaran and Naqvi with multiple overlapping counts: continuing financial-crimes enterprise, conspiracy to commit securities fraud, securities fraud, conspiracy to commit wire fraud, and wire fraud. This structure gives prosecutors multiple paths to conviction at trial and multiple bases for sentencing enhancements.
3. The timing. The DOJ unsealed the indictment just days after the FBI released its 2025 Internet Crime Report highlighting the growth of AI-enabled fraud. The message is clear: as AI becomes more central to the economy, federal enforcement against AI-related financial crimes will become more aggressive.

What the iLearning case means for the future of AI regulation

The iLearningEngines indictment is the leading edge of a broader enforcement wave that will likely include:

• More AI washing prosecutions targeting both public companies and private startups that defraud venture capitalists.
• SEC rulemaking requiring specific disclosures about AI usage, training data, and performance metrics.
• Civil lawsuits from investors who lost money in AI washing schemes, using the DOJ indictment as evidence in parallel class actions.
• International coordination as AI washing schemes cross borders, with the DOJ working alongside UK, EU, and Asian regulators.

For the average person, the lesson is straightforward: the AI label is not due diligence. A company can claim to be AI-powered, list on the Nasdaq, and reach a billion-dollar valuation, and still be a complete fraud. The only defense is verification. Check the filings. Read the short reports. Verify the pitch before you invest.

How to protect yourself

The iLearningEngines case offers painful lessons for anyone investing in AI companies, whether publicly traded stocks, private placements, or venture capital funds. AuthentiLens recommends five protections.

1. Treat unverified AI capability claims as marketing, not fact. Every AI company claims to have proprietary technology, breakthrough models, and unmatched performance. Few can prove it. Before you invest, ask to see the model, the architecture, the training data provenance, or a live technical demo with inputs you control. Companies with real AI have no trouble showing it. Companies that deflect, obscure, or offer only high-level marketing slides are giving you valuable information, just not the kind they intend.
2. Read SEC filings, not just press releases. For publicly traded AI companies, the truth is often hiding in plain sight in the 10-K and 10-Q filings. Pay special attention to:
   • Risk factors. Generic boilerplate is a red flag. Real AI companies disclose specific technical, regulatory, and competitive risks.
   • Management's Discussion and Analysis (MD&A). Look for vague explanations of revenue growth, or sudden changes in accounting methods.
   • Auditor changes. A company that switches auditors frequently, or replaces its auditor with a smaller, less reputable firm, is often trying to escape uncomfortable questions.
   • Related-party transactions. If a significant percentage of revenue comes from a handful of customers, and those customers have ties to company insiders, you are looking at the same pattern prosecutors allege in the iLearning case.
3. Read short-seller reports before you double down. Research reports from firms like Hindenburg Research, Muddy Waters, and Kerrisdale Capital have repeatedly identified fraud years before regulators acted. You do not have to agree with a short-seller's conclusions, but you should read their reports carefully before you increase your position in an AI company that has been publicly challenged. In the iLearning case, investors who read the August 2024 short report and sold their shares saved themselves from a total loss.
4. Be deeply suspicious of “AI investment opportunities” on social media. As the FBI's 2025 Internet Crime Report documented, AI-nexus investment scams caused $632 million in documented losses last year. If an “AI investment opportunity” appears in a direct message on Telegram, Discord, or WhatsApp, or in a paid influencer post on TikTok or Instagram, it is not an opportunity. It is a setup. Real investment opportunities exist within regulated channels: brokerages, SEC filings, and verifiable company websites.
5. Verify suspicious pitches before you wire a single dollar. If you receive an investment pitch that pressures you to act quickly or claims to offer “exclusive access” to an AI startup, stop. Copy the message text, save the video, or screenshot the pitch. Paste it into AuthentiLens. Our detection engine flags AI-generated content, impersonation signals, social-engineering language, and known scam patterns in seconds, before you reply, click, or pay.

If the only evidence that an AI company is real is its own marketing materials, you are not investing. You are hoping.